The present invention relates to an exhaust valve for an internal combustion engine, particularly a two-stroke crosshead engine, comprising a movable spindle with a valve disc which at its upper surface has a annular seat area of a material different from the base material of the valve disc, which seat area abuts a corresponding seat area on a stationary valve member in the closed position of the valve.
The development of exhaust valves for internal combustion engines has aimed for many years at extending the life and reliability of the valves. This has been done so far by manufacturing the valve spindles with a hot-corrosion-resistant material on the lower disc surface and a hard material in the seat area.
The seat area is particularly crucial for the reliability of the exhaust valve, as the valve has to close tightly to function correctly. It is well-known that the ability of the seat area to close tightly can be reduced by corrosion in a local area by a so-called burn through, where across the annular sealing surface a channel-shaped gutter emerges, through which hot gas flows when the valve is closed. Under unfortunate circumstances, this failure condition can arise and develop into a rejectable valve during less than 80 hours"" operation, which means that often it is not possible to discover the beginning failure at the usual overhaul. Therefore, a burn through in the valve seat may cause unplanned shut-downs. If the engine is a propulsion engine in a ship, the, condition may be initiated and develop into a failing valve during a single voyage between two ports, which may cause problems during the voyage and unintended expensive waiting time in port.
With a view to preventing burn throughs in the valve seat many different valve seat materials with ever increasing hardnesses have been developed over the years. to make the seat wear-resistant by means of the hardness and reduce the formation of dent marks. The dent marks are a condition for development of a burn through as the dents may create a. small leak through which hot gas flows. The hot gas can heat the material around the leak to a level of temperature where the gas with the aggressive components has a corrosive effect on the seat material so that the leak rapidly grows larger and the leakage flow of hot gas increases, which escalates the erosion. In addition to the hardness, seat materials have also developed towards a higher hot corrosion resistance to delay erosion after the occurrence of a small leak. The special requirements to the seat material and the deviant special requirements to material properties in other areas of the movable valve member necessitate a seat area of a material different from the base material of the valve disc, which also provides manufacturing advantages. A number of examples of known seat materials will be given below.
WO92/13179, for example, describes the use of the nickel-based alloy Alloy 50, the cobalt-based alloy Stellite 6 and a nickel-based alloy the most important alloy components of which are 20-24% Cr, 0.2-0.55% C and 4-7% Al. One object mentioned is that seat materials should be hard to reduce the formation of dent marks.
SE-B-422 388 describes a valve for an internal combustion engine having a base body made of a chromium-containing nickel alloy on which a chromium-containing cobalt alloy is deposited at a temperature exceeding 3000xc2x0 C., whereupon the body is exposed to mechanical treatment and aging at a temperature higher than the operating temperature. An object of this is to improve the corrosion resistance of the seat material and impart a high hardness to it.
DK-B-165125 teaches an exhaust valve for an internal combustion engine with a seat area of a corrosion-resistant facing alloy comprising 13-17% Cr, 2-6% Al, 0.1-8% Mo, 1.5-3.5% B, 0.5-3% Ti, 4-7% Co and a balance of Ni. High hardness of the seat material is desired.
U.S. Pat. No. 4,425,300 teaches a welded-on hardfacing alloy comprising 10-25% Cr, 3-15% Mo, 3-7% Si, 1-1.2% C, 1-30% Fe and a balance of Ni. The alloy is without porosity and has a hardness comparable with that of cobalt-based alloys.
EP-A-0529208 teaches a nickel and chromium containing hardfacing alloy for welding-on in the valve seat area in a car engine. The alloy contains 30-48% Ni, 1.5-15% W and/or 1.0-6.5% Mo and the balance is of at least 40% Cr. W and Mo have a solution-strengthening effect on the alloy. C can be added in amounts from 0.3 to 2.0% to increase the hardness by carbide formation, and B can be added in amounts from 0.1 to 1.5% to increase the hardness by chromium boride formation. Nb can be added in amounts from 1.0 to 4.0% for formation of hardness increasing intermetallic compounds as well as carbides and borides.
EP-A-0 521 821 teaches a valve made of NIMONIC 80A or NIMONIC 81, which is provided with a layer of INCONEL 625 or of INCONEL 671 in the seat area to impart to the seat a higher corrosion resistance than the NIMONIC base body. The publication mentions for the alloy INCONEL 671 that it only has to be welded on, while for the alloy INCONEL 625 it mentions that after the welding it contains a dendritic carbide structure and that the seat area therefore has to be hot-worked to homogenise the carbide distribution in the structure to improve corrosion resistance.
The book xe2x80x98Diesel engine combustion chamber materials for heavy fuel operationxe2x80x99 published in 1990 by The Institute of Marine Engineers, London, collects the experience gained for exhaust valve materials in a number of articles and provides recommendations as to how to design valves to achieve long life. Concerning valve seats the articles unanimously direct that the seat material has to have a high hardness and be of a material with a high resistance against hot corrosion. A number of different preferred materials for exhaust valves are described in Paper 7 of the book xe2x80x98The physical and mechanical properties of valve alloys and their use in component evaluation analysesxe2x80x99, including in its analysis of the mechanical properties of the materials a comparative table of the yield strength of the materials, seen to be below about 820 MPa.
It is desirable to prolong the life of the exhaust valve and particularly to reduce or avoid unpredictable and rapid development of burn throughs in the seat area of the valve. The Applicant has carried out tests with dent mark formation in seat materials and contrary to the established knowledge has established quite unexpectedly that the hardness of the seat material does not have any great influence on whether the dent marks emerge. The object of the. present invention is to provide seat materials that anticipate the mechanism leading to formation of dent marks, whereby the basic condition for occurrence of burn throughs is weakened or eliminated.
In view of this the exhaust valve according to the invention is characterised in that the seat area at the upper surface of the valve disc is made of a material which has a yield strength of at least 1000 MPa at a temperature of approximately 20xc2x0 C.
Dent marks are formed by particulate combustion residues, such as coke particles, which flow from the combustion chamber up through the valve and into the exhaust system while the exhaust valve is open. When the valve closes, the particles may get caught between the closing sealing surfaces on the valve seats.
From studies of numerous dent marks on valve spindles in operation it has been observed that new dent marks very rarely reach the upper closing rim, viz., the circumferential line at which the upper end of the stationary valve seat is brought into contact with the movable conical valve seat. In practice, the dents end about 0.5 mm away from the closing rim, which is without any immediate explanation, as a particle can also be expected to be caught in this area.
It has now been realized that the absence of dents immediately up to the closing rim is due to the fact that coke particles and other, even very hard particles are crushed to powder before the valve is completely closed. Part of the powder is blown away simultaneously with the crushing of the particles because the gas from the combustion chamber flows out through the gap between the closing sealing surfaces at approximately sonic velocity. The high gas velocity blows the powder near the closing rim away, and the absence of dents out to the rim shows that just about all particles getting caught between the sealing surfaces are pulverized. Even very thick particles are reduced in thickness by the crushing and blowing away of powder, and in practice the subsided piles of powder capable of forming the dent marks therefore have a highest thickness of 0.5 mm and a normal maximum thickness of 0.3-0.4 mm.
Especially within the most recent engine development where the maximum pressure may be 195 bar, the load on the lower surface of the disc may correspond to up. to 400 tons. When the exhaust valve is closed and the pressure in the combustion chamber rises to the maximum pressure, the sealing surfaces are pressed completely together around an enclosed powder pile. This cannot be prevented, no matter how hard the seats are made.
When the combustion of the fuel commences and the pressure in the cylinder and the load on the valve disc increase, the enclosed powder pile starts wandering into the two sealing surfaces and at the same time the seat materials are elastically deformed. During this elastic deformation the surface pressure between the powder pile and the sealing surfaces rises, which usually makes the powder pile deform into a larger area. If the powder pile is sufficiently thick, the elastic deformation continues until the pressure in the contact area of the powder pile reaches the yield strength of the seat material of the lowest yield strength, whereupon this seat material is plastically deformed and formation of the dent mark commences. The plastic deformation may result in an increase of the yield strength owing to deformation hardening. If the two seat materials in the local area around the powder pile thus achieve uniform yield strengths, the powder pile starts plastically deforming the other seat material as well.
If the formation of dent marks is to be countered, this, as mentioned above, cannot be done by making seat materials harder, instead they have to be made resilient, which is obtained by manufacturing the seat areas with a high yield strength. The higher yield strength provides a double effect. Firstly, the seat material with the higher yield strength may be exposed to a higher elastic strain and thus absorb a thicker powder pile before plastic deformation occurs. The second essential effect is associated with the surface nature of the sealing surfaces in the areas facing the powder pile. The dent profile formed by the elastic deformation is even and smooth and promotes the distribution of the powder pile to a larger diameter, which partly reduces the thickness of the powder pile, partly reduces stresses in the contact area following from the greater contact area. At the transition from elastic deformation to plastic deformation a deeper and more irregular dent profile is rapidly created which will unsuitably anchor the powder pile and thus have a preventive effect on a further advantageous enlargement of the diameter of the pile.
Tests have shown that in an exhaust valve a powder pile of a thickness of about 0.14 mm can be absorbed between two seat areas of materials with a lower limit for the yield strength of 1000 MPa without any plastic deformation of the sealing surfaces. A large proportion of the particles caught between the seat surfaces will be crushed to a thickness of about 0.15 mm. The exhaust valve according to the invention prevents a noticeable proportion of the particles from forming dent marks because the seat surface merely springs back to its original shape when the valve opens, and at the same time the remains of the crushed particle are blown away from the seat surfaces.
In consideration of an increase of the elastic properties of the seat area, it is preferred that the seat area material has a yield strength of at least 1100 MPa, preferably at least 1200 MPa. Young""s modulus for the current seat material is substantially unchanged at increasing yield strengths, which gives an approximately linear correlation between yield strength and the highest elastic strain. It appears from the above comments that a seat material with a yield strength of 2500 MPa or more would be ideal because it could absorb the powder piles of the normally most frequently occurring pile thicknesses purely by elastic deformation. However, at present suitable materials with such a high yield strength are not at hand. It will appear from the below description that some of the seat materials available today can be manufactured in a manner that raises the yield strength to at least 1100 MPa. All other things being equal, this 10% increase in yield strength will result in at least a 10% reduction of the depth of any dent marks. For most types of particles, the suitable limit of 1200 MPa is sufficiently high to provide a noticeable reduction of the pile thickness and consequently may result in a reduction of the dent mark depths of up to 30%, but at the same time the number of possible materials is narrowed down. This also applies to seat materials with a yield strength of at least 1300 MPa.
In an especially preferred embodiment the seat area material has a yield strength of at least 1400 MPa. This yield strength is almost double the yield strength of the seat materials used at present, and based on the present understanding of the mechanism of dent mark formation the material with this high yield strength is presumed to largely eliminate problems with seat area burn throughs. The depth of the few dent marks that can be formed in this seat material will be too small for leakage gas to flow through the dent mark in sufficiently large quantities for the seat material to be heated to a temperature where hot corrosion becomes effective.
In one embodiment the seat areas on the stationary member and the valve disc, respectively, have substantially the same yield strength at the operating temperatures of the seat areas. The largely uniform yield strengths of the two seat materials result in approximately the same manner of deformation of both sealing surfaces when the powder pile is pressed into the surfaces, which reduces the resulting plastic deformation in each of the surfaces. The stationary seat area is colder than the seat area on the spindle, which means that the spindle seat material should have the higher yield strength at about 20xc2x0 C. in view of the fact that the yield strength for many materials drops at increasing temperatures. This embodiment is especially advantageous if the stationary seat area is made of a hot-corrosion-resistant material.
If the stationary seat area is of hardened steel or cast iron, the seat area on the stationary member preferably has a substantially higher yield strength than the seat area on the valve disc at the operating temperatures of the seat areas. With this design any dent marks will be formed on the valve spindle. This provides two advantages. Firstly the seat area on the spindle is normally made of a hot-corrosion-resistant material so that any dent mark will find it more difficult to develop into a burn through than if the dent was located on the stationary member. Secondly the spindle rotates so that at each valve closure the dent will be located at a new position on the stationary sealing surface, the heat influence thus being distributed on the stationary seat area.
The following will describe different materials usable according to the invention as valve seat materials. It should be noted that NIMONIC and INCONEL are proprietary trademarks of INCO Alloys, and that Udimet is a proprietary trademark of Special Metals Inc.
The seat area material may be a nickel-based chromium-containing alloy comprising in terms of per cent by weight at least 10% of solution-strengthening components, such as Mo, W, Co, Hf, Fe and/or Cr, where the alloy is welded on to the valve disc and then the yield strength of the alloy has been increased. to a value higher than the above lower limit by cold-working of the material at a temperature lower than or around the recrystallisation temperature of the alloy. The following can be mentioned as examples of alloys of this type: IN 625 has a yield strength after welding of approximately 450 MPa, but after at least 27% cold-working the yield strength is approximately 1000 MPa, and after 40% cold-working about approximately 1100 MPa. IN 671 has a yield strength of approximately 490 MPa in a welded condition, and cold-working of between 30 and 40% may bring the yield strength above 1000 MPa. After the welding, IN 690 has a yield strength of approximately 500 MPa, and after cold-working of approximately 45% the yield strength of this alloy has been increased to approximately 1035 MPa. IN 718-like alloys also have a yield strength of approximately 500 MPa after the welding, and after cold-working of at least 35% the yield strength has been brought to just over 1000 MPa. However, not all IN 718-like alloys exhibit a strong increase of the yield strength at cold-working or heat treatment, which will be described in further detail below.
For the alloys containing Nb and/or Ta a further increase of the yield strength of the alloy can be achieved after cold-working by means of a precipitation-hardening heat treatment. This also applies to the alloys containing Al and Ti, but they normally require a fine adjustment of these two components and further suffer from the minor disadvantage that after welding it may be necessary to carry out solution annealing with subsequent heat treatment to enable the cold-working, Al and Ti having a precipitation-hardening effect already at welding.
Alternatively the seat area material may be a nickel-based chromium-containing alloy containing Nb and/or Ta, the alloy being welded on to the valve disc whereupon its yield strength has been increased to a value higher than said lower limit by means of a precipitation-hardening heat treatment. An example of such an alloy capable of achieving a high yield strength without cold-working is Rene 220. After welding, this alloy has a low yield strength, but at a suitable heat treatment the yield strength can easily in terms of manufacturing be brought substantially above 1000 MPa. NIMONIC Alloy PK31 and IN 718-like alloys can be given yield strengths of substantially above 1000 MPa by heat treatment without cold-working.
A further alternative that involves no cold-working either, is that the seat area material is a nickel-based chromium-containing alloy containing in terms of per cent by weight at least 10% of solution-strengthening components, such as Mo, W, Co, Hf, Fe and/or Cr, and precipitation-hardening components, such as Nb, Ta, Al and/or Ti, and that the alloy is welded on to the valve disc and then its yield strength is increased to a value higher than said lower limit by means of a precipitation-hardening heat treatment. As these alloys contain solution-strengthening components they have a tendency towards increasing the yield strength if in practice they are exposed to plastic deformation by a powder pile.
In another embodiment the seat area material is a nickel-based chromium-containing alloy including at least one component selected from among Co, Mo, Hf, Fe, W, Ti, Nb, Ta, Al, and at least the seat area is manufactured by means of a HIP process, possibly with a subsequent heat treatment to provide controlled precipitation hardening, typically solution annealing followed by quenching and precipitation hardening. Among especially applicable alloys can be mentioned IN 100, which has a yield strength of approximately 1300 MPa at approximately 20xc2x0 C. after the HIP process and is further especially advantageous in that the yield strength is maintained at a very high level at the operating temperature of the spindle, the yield strength being approximately 1285 MPa at 650xc2x0 C. After the HIP process, Merl 76 has a yield strength of approximately 1200 MPa, and Udimet 700 has a correspondingly high yield strength. Rene 95 is also suitable and after the HIP process has a yield strength of approximately 1230 MPa dropping to approximately 1160 MPa at 500xc2x0 C. The alloy NIMONIC Alloy 105 can also be used, possibly with a minor modification of the components forming carbonitride compounds and oxide compounds which, after the HIP process, can form coherent chains of brittle compounds, so-called PPBs (Prior Particle Boundaries). To the extent that these alloys contain solution-strengthening components the yield strength can be further increased by cold-working. The HIP process can also be supplemented with forging and extrusion processes. As an alternative to the HIP process, other powder metallurgical compacting processes can also be used with the above seat materials.
In yet another embodiment the seat area material is a nickel-based chromium-containing alloy including at least one component selected from among Co, Mo, W, Hf, Fe, Ti, Nb, Ta, Al, the seat area being manufactured by means of either casting or powder metallurgical application followed by thermo-mechanical forging, rolling or beating at a temperature lower than or around the recrystallisation temperature of the alloy and with a degree of deformation of the seat area increasing the yield strength of its material to a value higher than said lower limit. The powder metallurgical application may, for example, be thermal spraying-on of particulate or powdery starting material on a spindle base body, and the thermo-mechanical forging may comprise cold-working of the sprayed-on material. Preferably the cold-working takes place at a suitably raised temperature so as to avoid precipitation hardening in an extent that may bother the deformation process. The seat area may, for example, be made of an IN 718-like alloy that may have been exposed to a degree of deformation of at least 35%. The seat area may also be made of INCONEL Alloy X-750 which has been hot-worked and precipitation-hardened to a yield strength of approximately 1110 MPa. If the alloy contains precipitation-hardening components of the above type, it is further possible to increase the yield strength further through a precipitation-hardening heat treatment.
Especially advantageous alloys for the seat area material comprise 10-25% Cr, at the most 25% Co, at the most 10% Mo+W, at the most 11% Nb, at the most 20% Ta, at the most 3% Ti, at the most 0.55% Al, at the most 0.3% C, at the most 1% Si, at the most 0.015% P, at the most 0.015% S, at the most 3% Mn, at the most 25% Fe and a balance of Ni, and preferably the components Al, Ti and Ni are limited to up to 0.5% Al, 0.7-3% Ti and 52-57% Ni, the content of Nb+Ta/2 suitably being at least 3%.
The choice of alloy and the consequent manufacturing process may be influenced by the size of the exhaust valve, as a cold-working of many per cent may require strong tools when the valve disc is large, for example with the external diameter being in the interval from 130 mm to 500 mm.
The present invention also relates to the use of a nickel-based chromium-containing alloy with a yield strength of at least 1000 MPa at approximately 20xc2x0 C. as a dent mark limiting or preventive material in an annular seat area at the upper surface of a movable valve disc in an exhaust valve for an internal combustion engine, particularly a two-stroke crosshead engine, the seat area being made of an alloy different from the base material of the valve disc, and abutting a corresponding seat area on a stationary valve member when the valve is closed. The special advantages of using such a dent mark limiting material appear from the above description.